CN107274298A - A kind of agricultural product price fluctuation method for early warning and system - Google Patents

Abstract

The present invention discloses a kind of agricultural product price fluctuation method for early warning and system.Method includes：Historical price fluctuation to the agricultural product of species to be measured in region to be measured is sampled, and obtains sample data；Sample data is analyzed using Density Estimator method, the probability density function of agricultural product price fluctuation is obtained；The price fluctuation value in the range of the corresponding default price fluctuation of each police's degree is obtained from sample data；Least square computing is carried out to price fluctuation value, the corresponding price fluctuation point position probability of each alert degree is obtained；According to the price fluctuation scope of probability density function police's degree different with price fluctuation point position determine the probability；The agricultural product price undulating value of species to be measured in region to be measured and price fluctuation scope are contrasted, when agricultural product price undulating value is in the range of price fluctuation, the early warning of correspondence police's degree is sent.Method and system disclosed by the invention, can carry out accurate early warning to the price fluctuation of agricultural product, improve the objectivity of agricultural product price fluction analysis.

Description

A method and system for early warning of price fluctuations of agricultural products

technical field

The invention relates to the technical field of price monitoring of agricultural products, in particular to a method and system for early warning of price fluctuations of agricultural products.

Background technique

Agricultural products are an industrial form that combines natural reproduction and social reproduction, and play a vital role in the development of agriculture and rural economy. Therefore, analyzing the price fluctuations of agricultural products is of great significance for the government to carry out macro-control and farmers to designate production strategies.

The price of agricultural products is affected by the spatial distribution of production, the time distribution of product consumption, and the lag of market signals and policy effects. This makes the price fluctuation of agricultural products a relatively complicated process of change. At present, most of the means of analyzing the price fluctuation of agricultural products at home and abroad are directly judged by experts. However, the means of expert judgment is often limited by the experience of experts, and the analysis results are relatively subjective.

Contents of the invention

The object of the present invention is to provide a method and system for early warning of price fluctuations of agricultural products, which can accurately warn the price fluctuations of agricultural products and improve the objectivity of the analysis of price fluctuations of agricultural products.

To achieve the above object, the present invention provides the following scheme:

An early warning method for price fluctuations of agricultural products, comprising:

Sampling the historical price fluctuations of the agricultural products of the type to be tested in the area to be tested to obtain sample data;

Using a kernel density estimation method to analyze the sample data to obtain a probability density function of price fluctuations of agricultural products; the probability density function is a function reflecting the possible probability of the price fluctuation value of the agricultural products;

Acquiring the price fluctuation value within the preset price fluctuation range corresponding to each warning degree from the sample data; the warning degree represents the degree of warning, and the warning degree is proportional to the price fluctuation value of the agricultural product;

Carrying out the least squares operation on the price fluctuation value to obtain the price fluctuation quantile probability corresponding to each of the warning degrees; the price fluctuation quantile probability is a probability smaller than the price fluctuation value corresponding to the warning degree;

Determine price fluctuation ranges with different warning degrees according to the probability density function and the price fluctuation quantile probability;

The price fluctuation value of the agricultural product to be tested in the area to be tested is compared with the price fluctuation range, and when the price fluctuation value of the agricultural product is within the price fluctuation range, an early warning corresponding to the warning degree is issued.

Optionally, the analysis of the sample data by using the kernel density estimation method to obtain the probability density function of agricultural product price fluctuations specifically includes:

Using a Gaussian kernel function as the kernel function of the kernel density estimation method, analyzing the sample data to obtain a probability density function;

Select the window width of the probability density function to determine the optimal window width; the optimal window width is the window width that minimizes the deviation of the probability density function relative to the sample data; the window width is the Parameters of the probability density function.

The invention also discloses an early warning system for price fluctuations of agricultural products, including:

The sampling module is used to sample historical price fluctuations of agricultural products of the type to be tested in the area to be tested to obtain sample data;

The kernel density estimation module is used to analyze the sample data by using the kernel density estimation method to obtain the probability density function of the price fluctuation of agricultural products; the probability density function is a function reflecting the possible probability of the price fluctuation value of the agricultural products ;

The warning degree fluctuation value acquisition module is used to obtain the price fluctuation value within the preset price fluctuation range corresponding to each warning degree from the sample data; the warning degree indicates the degree of warning, and the warning degree is related to the price of the agricultural product The fluctuation value is proportional to;

The quantile calculation module is used to perform the least squares operation on the price fluctuation value to obtain the price fluctuation quantile probability corresponding to each of the warning degrees; the price fluctuation quantile probability is the ratio corresponding to the warning degree The probability that the price fluctuation value of is small;

A warning range determination module, configured to determine price fluctuation ranges with different warning degrees according to the probability density function and the price fluctuation quantile probability;

The early warning module is used to compare the price fluctuation value of the agricultural product of the type to be tested in the area to be tested with the price fluctuation range, and when the price fluctuation value of the agricultural product is within the price fluctuation range, issue a corresponding alarm early warning.

Optionally, the kernel density estimation module specifically includes:

A probability density function determination unit, configured to use a Gaussian kernel function as the kernel function of the kernel density estimation method, analyze the sample data, and obtain a probability density function;

The window width selection unit is used to select the window width of the probability density function and determine the optimal window width; the optimal window width is the window width that minimizes the deviation of the probability density function relative to the sample data; The window width is a parameter of the probability density function.

According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects: the agricultural product price fluctuation early warning method and system of the present invention can accurately obtain the price of a specific type of agricultural product in a specific area by sampling and analyzing the historical price fluctuation of agricultural products. Price fluctuation rules, so as to determine the range of price fluctuations with different warning degrees, can accurately warn the price fluctuations of agricultural products. At the same time, because the method and system of the present invention are analyzed according to the region and the type of agricultural products, it can be applied to the price warning of various agricultural products, has universal applicability and pertinence, and can improve the accuracy while improving the application range of the method. Moreover, the method of the present invention is based on historical data and obtained through objective analysis. Compared with the existing means of directly judging by experts, the objectivity of agricultural product price fluctuation analysis is improved.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required in the embodiments. Obviously, the accompanying drawings in the following description are only some of the present invention. Embodiments, for those of ordinary skill in the art, other drawings can also be obtained according to these drawings without paying creative labor.

Fig. 1 is the method flowchart of Embodiment 1 of the method for early warning of price fluctuations of agricultural products of the present invention;

Fig. 2 is a system structure diagram of an embodiment of the agricultural product price fluctuation early warning system of the present invention.

detailed description

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

In order to make the above objects, features and advantages of the present invention more comprehensible, the present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

Embodiment one:

FIG. 1 is a flow chart of Embodiment 1 of the method for early warning of price fluctuations of agricultural products in the present invention.

See Figure 1, the early warning method for price fluctuations of agricultural products, including:

Step 101: Sampling the historical price fluctuations of agricultural products of the type to be tested in the area to be tested to obtain sample data;

Step 102: Analyze the sample data using the kernel density estimation method to obtain the probability density function of the price fluctuation of agricultural products; the probability density function is a function reflecting the possible probability of the price fluctuation value of the agricultural product; this step is specifically include:

Using a Gaussian kernel function as the kernel function of the kernel density estimation method, analyzing the sample data to obtain a probability density function;

Select the window width of the probability density function to determine the optimal window width; the optimal window width is the window width that minimizes the deviation of the probability density function relative to the sample data; the window width is the parameters of the probability density function;

Step 103: Obtain the price fluctuation value within the preset price fluctuation range corresponding to each warning degree from the sample data; the warning degree indicates the degree of warning, and the warning degree is proportional to the price fluctuation value of the agricultural product;

Step 104: Perform the least squares operation on the price fluctuation value to obtain the price fluctuation quantile probability corresponding to each of the warning degrees; the price fluctuation quantile probability is smaller than the price fluctuation value corresponding to the warning degree The probability;

Step 105: Determine the range of price fluctuations with different warning degrees according to the probability density function and the quantile probability of price fluctuations;

Step 106: Compare the price fluctuation value of the agricultural product to be tested in the area to be tested with the price fluctuation range, and when the price fluctuation value of the agricultural product is within the price fluctuation range, issue an early warning corresponding to the warning degree .

The agricultural product price fluctuation early warning method of the present invention accurately obtains the price fluctuation law of a specific type of agricultural product in a specific area by sampling and analyzing the historical price fluctuation of agricultural products, thereby determining the price fluctuation range of different warning degrees, and can predict the price of agricultural products. Accurate early warning of fluctuations. At the same time, because the method of the present invention is analyzed according to the region and the type of agricultural products, it can be applied to the price warning of various agricultural products, has universal applicability and pertinence, and can improve the accuracy while increasing the application range of the method. Moreover, the method of the present invention is based on historical data and obtained through objective analysis. Compared with the existing means of directly judging by experts, the objectivity of agricultural product price fluctuation analysis is improved.

Embodiment two:

An early warning method for price fluctuations of agricultural products, comprising:

Sampling the historical price fluctuations of the agricultural products of the type to be tested in the area to be tested to obtain sample data.

Analyzing the sample data using a kernel density estimation method to obtain a probability density function of price fluctuations of agricultural products;

The probability density function is expressed as:

in is the probability density function, n is the quantity of the sample data, and i is the sequence number of the sample data; Xi is the _i -th sample data, K(x) is a kernel function, x is the variable of the probability density function, and h is window width.

Select Gaussian kernel function as kernel function in the present embodiment, then described kernel function K (x) is:

The deviation and error of the probability density function obtained according to the kernel density estimation method change in different directions with the change of the window width h. The curve is not smooth, so the probability density function is sensitive to the choice of window width h.

In order to make the probability density function closer to the real function, this embodiment adopts the mean square error (MSE, Mean Squared Error) criterion to select the window width, so that the probability density function is closer to the real function f(x). The mean square error criterion requires that the mean square error be minimum. The formula for calculating the mean square error is:

in represents the probability density function Variance, Indicates the deviation of the probability density function from the true function f(x). expectations, expectations. o(h ⁴ ) is the infinitesimal quantity of h ⁴ , for an infinitesimal amount of .

When calculating the window width, in this embodiment, the least squares cross-validation method is used to calculate the optimal window width. The least squares cross-validation method does not need to make any assumptions on the probability density function, and can obtain the optimal window width directly from actual data. In the least squares cross-validation method, the window width when the integrated square error (ISE, Integrated square error) of the probability density function is minimized is the optimal window width. which is:

in Indicates the window width when the integral square difference is the smallest, h _opt is the optimal window width;

but

Among them, X _j is the jth sample data.

The probability density function can be determined after the window width is determined by the above method.

After calculating the probability density function, the p-quantile method is used to calculate the fluctuation range of agricultural product prices corresponding to each warning degree, specifically:

Acquiring the price fluctuation value within the preset price fluctuation range corresponding to each warning degree from the sample data; the warning degree represents the degree of warning, and the warning degree is proportional to the price fluctuation value of the agricultural product;

The least squares operation is performed on the price fluctuation value to obtain the price fluctuation quantile probability corresponding to each of the warning degrees; the price fluctuation quantile probability is a probability smaller than the price fluctuation value corresponding to the warning degree.

The setting of quantile probabilities of price fluctuations requires a combination of qualitative and quantitative methods, mainly following the following principles: First, the importance and comprehensiveness of economic meanings. For example, the distribution of the probability density function of fruit and vegetable products is obviously different from that of meat products, so different price fluctuation quantile probabilities should be set for different types of agricultural products; at the same time, as the regional output decreases, the corresponding regional price fluctuations will also be more significant , so the setting of the quantile probability of price fluctuations in different regions should also be different. The second is the measurement ability of indicators. The response of each alarm level to the price change of the overall agricultural product market needs to be sensitive and reliable, as well as relatively stable. The third is the timeliness of the indicators. The indicators are not static, but are adjusted, supplemented and modified over time to meet the needs of market early warning.

Therefore, the quantile probability of price fluctuation can be obtained by using the least square method, namely:

Among them, p ^(m) is the price fluctuation quantile probability of the mth warning level, x _i is the price fluctuation value within the preset price fluctuation range corresponding to the mth warning level; x ^(m) is the mth warning level The expected quantile corresponding to the quantile probability of price fluctuation of .

The present invention converts the discontinuous sample data into a continuous function, and uses the continuous function to calculate the price fluctuation quantile probability of each warning level, thereby accurately calculating the specific value of the quantile corresponding to each price fluctuation quantile probability, It makes the price fluctuation range corresponding to each warning level more accurate, and improves the accuracy of early warning.

After calculating the price fluctuation quantile probability p ^(m) corresponding to each alert degree, the quantile corresponding to each price fluctuation quantile probability can be calculated according to the probability density function The quantile is the price fluctuation value corresponding to the price fluctuation quantile probability in the probability density function; the specific calculation formula is:

in, is the quantile corresponding to the price fluctuation quantile probability p ^(m) ;

Different warning degrees correspond to different quantiles, and the price fluctuation range of each warning degree can be calculated according to the quantiles.

In this embodiment, there are four warning levels, which respectively correspond to four warning lights of different colors when calling the police. The colors of the four warning lights are respectively red, orange, yellow and blue according to the degree of warning from high to low. Among them, the blue light indicates that the price fluctuation is slightly high, the yellow light indicates that the price fluctuation is high, the orange light indicates that the price fluctuation is too high, and the red light indicates that the price fluctuation is extremely high. When the price fluctuation value of agricultural products is within the price fluctuation range corresponding to a certain warning degree, the warning light of the corresponding color will be lit.

Embodiment three:

Fig. 2 is a system structure diagram of an embodiment of the agricultural product price fluctuation early warning system of the present invention.

See Figure 2, the agricultural product price fluctuation early warning system includes:

The sampling module 201 is used to sample the historical price fluctuations of the agricultural products of the type to be tested in the area to be tested to obtain sample data;

The kernel density estimation module 202 is configured to use a kernel density estimation method to analyze the sample data to obtain a probability density function of price fluctuations of agricultural products; the probability density function reflects the probability of possible occurrence of the price fluctuation value of the agricultural products function;

The warning degree fluctuation value acquisition module 203 is used to obtain the price fluctuation value within the preset price fluctuation range corresponding to each warning degree from the sample data; the warning degree indicates the degree of warning, and the warning degree and the agricultural product The price fluctuation value is proportional;

The quantile calculation module 204 is used to perform the least squares operation on the price fluctuation value to obtain the price fluctuation quantile probability corresponding to each of the warning degrees; The probability that the corresponding price fluctuation value is small;

A warning degree range determination module 205, configured to determine price fluctuation ranges with different warning degrees according to the probability density function and the price fluctuation quantile probability;

The early warning module 206 is used to compare the price fluctuation value of the agricultural product of the type to be tested in the area to be tested with the price fluctuation range, and when the price fluctuation value of the agricultural product is within the price fluctuation range, send a corresponding alarm degree of warning.

The kernel density estimation module 202 specifically includes:

A probability density function determination unit, configured to use a Gaussian kernel function as the kernel function of the kernel density estimation method, analyze the sample data, and obtain a probability density function;

The window width selection unit is used to select the window width of the probability density function and determine the optimal window width; the optimal window width is the window width that minimizes the deviation of the probability density function relative to the sample data; The window width is a parameter of the probability density function.

The agricultural product price fluctuation early warning system of the present invention accurately obtains the price fluctuation law of a specific type of agricultural product in a specific area by sampling and analyzing the historical price fluctuation of agricultural products, thereby determining the price fluctuation range of different warning degrees, and can predict the price of agricultural products. Accurate early warning of fluctuations. Simultaneously, because the system of the present invention analyzes according to regions and types of agricultural products, it can be applied to price warnings of various agricultural products, has universal applicability and pertinence, and can improve accuracy while increasing the application range of the method. Moreover, the method of the present invention is based on historical data and obtained through objective analysis. Compared with the existing means of directly judging by experts, the objectivity of agricultural product price fluctuation analysis is improved.

Each embodiment in this specification is described in a progressive manner, each embodiment focuses on the difference from other embodiments, and the same and similar parts of each embodiment can be referred to each other. As for the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and for the related information, please refer to the description of the method part.

In this paper, specific examples have been used to illustrate the principle and implementation of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; meanwhile, for those of ordinary skill in the art, according to the present invention Thoughts, there will be changes in specific implementation methods and application ranges. In summary, the contents of this specification should not be construed as limiting the present invention.

Claims (4)

1. An agricultural product price fluctuation early warning method is characterized by comprising the following steps:

sampling historical price fluctuation of agricultural products of a to-be-detected type in a to-be-detected area to obtain sample data;

analyzing the sample data by using a nuclear density estimation method to obtain a probability density function of agricultural product price fluctuation; the probability density function is a function reflecting a probability of a possible occurrence of a price fluctuation value of the agricultural product;

acquiring price fluctuation values in a preset price fluctuation range corresponding to each alarm degree from the sample data; the alarm degree is in direct proportion to the price fluctuation value of the agricultural product;

performing least square operation on the price fluctuation value to obtain price fluctuation quantile probability corresponding to each alarm degree; the price fluctuation quantile probability is smaller than the price fluctuation value corresponding to the alarm degree;

determining price fluctuation ranges with different degrees of alarm according to the probability density function and the price fluctuation quantile probability;

and comparing the price fluctuation value of the agricultural product of the type to be detected in the area to be detected with the price fluctuation range, and sending out an early warning corresponding to the alarm degree when the price fluctuation value of the agricultural product is in the price fluctuation range.

2. The agricultural product price fluctuation early warning method according to claim 1, wherein the analyzing the sample data by using a kernel density estimation method to obtain a probability density function of agricultural product price fluctuation specifically comprises:

taking a Gaussian kernel function as a kernel function of the kernel density estimation method, and analyzing the sample data to obtain a probability density function;

selecting the window width of the probability density function, and determining the optimal window width; the optimal window width is the window width which enables the deviation of the probability density function relative to the sample data to be minimum; the window width is a parameter of the probability density function.

3. An agricultural product price fluctuation early warning system, comprising:

the sampling module is used for sampling the historical price fluctuation of the agricultural products of the type to be detected in the region to be detected to obtain sample data;

the nuclear density estimation module is used for analyzing the sample data by using a nuclear density estimation method to obtain a probability density function of agricultural product price fluctuation; the probability density function is a function reflecting a probability of a possible occurrence of a price fluctuation value of the agricultural product;

the system comprises a sample data acquisition module, a price fluctuation value acquisition module and a price fluctuation value acquisition module, wherein the sample data acquisition module is used for acquiring price fluctuation values in a preset price fluctuation range corresponding to various polices from the sample data; the alarm degree is in direct proportion to the price fluctuation value of the agricultural product;

the quantile calculation module is used for performing least square operation on the price fluctuation value to obtain price fluctuation quantile probability corresponding to each alarm degree; the price fluctuation quantile probability is smaller than the price fluctuation value corresponding to the alarm degree;

the alarm degree range determining module is used for determining price fluctuation ranges with different alarm degrees according to the probability density function and the price fluctuation quantile probability;

and the early warning module is used for comparing the price fluctuation value of the agricultural product of the type to be detected in the area to be detected with the price fluctuation range, and sending out early warning corresponding to the alarm degree when the price fluctuation value of the agricultural product is in the price fluctuation range.

4. The agricultural product price fluctuation early warning system according to claim 3, wherein the kernel density estimation module specifically comprises:

a probability density function determining unit, configured to analyze the sample data to obtain a probability density function, with a gaussian kernel function as a kernel function of the kernel density estimation method;

the window width selection unit is used for selecting the window width of the probability density function and determining the optimal window width; the optimal window width is the window width which enables the deviation of the probability density function relative to the sample data to be minimum; the window width is a parameter of the probability density function.